Technical description
A small group of unauthorized users gained access to Anthropic's Mythos AI model through a private online forum on the same day the company announced limited enterprise testing.
Attack vector
Unauthorized access through unspecified means to a restricted AI model designed for cybersecurity research and vulnerability discovery.
Affected systems
Anthropic Mythos AI model, described as capable of enabling dangerous cyberattacks and finding vulnerabilities at expert researcher level.
Mitigation
Anthropic is restricting access to the model to help governments, tech giants, and banks secure their systems ahead of broader release.